Fabrication and corrosion behavior of TiO2 nanotubes

0 downloads 0 Views 969KB Size Report
Jul 13, 2017 - Jianfang Li, Xiaojing He, Ruiqiang Hang, Xiaobo Huang, Xiangyu Zhang. ⁎. , Bin Tang. College of Materials Science and Engineering, Taiyuan ...
Ceramics International 43 (2017) 13683–13688

Contents lists available at ScienceDirect

Ceramics International journal homepage: www.elsevier.com/locate/ceramint

Fabrication and corrosion behavior of TiO2 nanotubes on AZ91D magnesium alloy

MARK



Jianfang Li, Xiaojing He, Ruiqiang Hang, Xiaobo Huang, Xiangyu Zhang , Bin Tang College of Materials Science and Engineering, Taiyuan University of Technology, Taiyuan 030024, China

A R T I C L E I N F O

A BS T RAC T

Keywords: AZ91D magnesium alloy TiO2 nanotube array Anodization Corrosion resistance

The major drawback of magnesium alloys in biomedical applications is the rapid degradation rate and the lack of biological activity. In this study, TiO2 nanotubes were fabricated on the surface of AZ91D magnesium alloy (TiO2-Mg) to overcome such limitations. The corrosion behavior of TiO2-Mg nanotubes was studied in simulated body fluid solution using open circuit potentials (OCP), electrochemical impedance spectroscopy (EIS) and potentiodynamic polarization tests. The high polarization resistance and open circuit potentials of TiO2-Mg nanotubes indicate the formation of highly stable TiO2 layer in simulated body fluid than that of titanium layer on magnesium alloy (Ti-Mg). TiO2 nanotubes on AZ91D magnesium alloy (AZ91D) can effectively decrease the degradation rate of magnesium alloy, thus can be further applied in orthopedic implants.

1. Introduction Magnesium (Mg) alloys are potential biomaterials in orthopedic implants on account of their good mechanical strength, biodegradable and non-toxicity [1–3]. Mg, an essential element of the human body, involved in the synthesis of protein and nucleic acid, regulation of cell cycle, adhesion of osteoblastic cells and bone growth [2,4,5]. However, the very fast degradation rate and low corrosion resistance in a biological environment have limited its extensive biomedical application [3,4,6]. Accordingly, Mg-based implants with improved corrosion resistance are highly necessary. Moreover, a large number of dissolved Mg can generate hydrogen gas and alkalization of body fluid thereby resulting in inflammation of the surrounding tissue [7,8]. Once occurred, loss of mechanical integrity is extremely difficult to support the weight. Thus, the effective surface modification of implants is necessary to slow down Mg corrosion. Being coated by comparative inert metal or its oxides is proved to be a potential method to improve the electrochemistry character. Among the various processing methods, pulsed direct current magnetron sputtering processing is an effective approach to the deposition of metallic and ceramic coatings onto various substrates forming uniform and compact film [9]. Titanium (Ti) has been widely used as orthopedic implant materials due to the excellent biocompatibility and corrosion resistance [10–12]. TiO2 nanotubes (NTs) prepared by electrochemical anodization of Ti has attracted much attention due to the controlling dimensions, large specific surface area and vertically aligned to the surface of the substrate [13,14]. Moreover, NTs with proper dimen-



Corresponding author. E-mail address: [email protected] (X. Zhang).

http://dx.doi.org/10.1016/j.ceramint.2017.07.079 Received 2 June 2017; Received in revised form 4 July 2017; Accepted 11 July 2017 Available online 13 July 2017 0272-8842/ © 2017 Elsevier Ltd and Techna Group S.r.l. All rights reserved.

sions have been reported to promote the corrosion resistance and biological activity [15–18]. In the present work, Ti films were deposited on AZ91D by magnetron pulsed direct current sputtering first, and then TiO2-Mg NTs were prepared by anodization. The different voltage and time of TiO2-Mg NTs were investigated. The OCP, EIS and potentiodynamic polarization were performed to investigate the corrosion resistance of TiO2-Mg NTs in simulated body fluid. 2. Materials and methods 2.1. Preparation of Ti coatings An AZ91D magnesium alloy with chemical composition (wt%) of 9% Al, 0.5% Zn, 0.15% Mn, 0.032% Si, 0.01% Cu, 0.002% Fe and balanced Mg was used. The AZ91D with dimensions of 10 mm × 10 mm × 3 mm was successively grinded with emery papers. And then these samples were respectively cleaned by acetone, ethanol, and distilled water for 10 min and dried in cool air. Ti coatings were deposited on the magnesium alloy by pulsed DC magnetron sputtering using pure Ti target. Before deposition, the AZ91D and the Ti target were sputtered for 30 min at a bias of 800 V, a duty factor of 40%, a pulse frequency of 60 kHz, and a working pressure of 5.0 Pa. After that, the Ti coatings were deposited according to the follow deposition parameters: target power, 280 W; working pressure, 0.8 Pa; substrate bias, 80 V; pulse frequency, 60 kHz; duty factor, 60%; duration, 3 h.

Ceramics International 43 (2017) 13683–13688

J. Li et al.

Fig. 1. Characterization of TiO2-Mg NTs (a, b) and TiO2 NTs (c, d) at different voltages for 1 h. (a, c) the surface SEM images of the anodized samples; (b, d) histogram of the nanotube diameter; (e) current density-time curves acquired from the samples at 30 V.

2.2. Fabrication of TiO2-Mg NTs TiO2-Mg NTs were fabricated by anodizing as-deposited Ti coatings. The electrolyte was an ethylene glycol solution containing 0.6 wt% NH4F and 2 vol% H2O. Anodization was investigated in a conventional two-electrode system with the Ti-Mg samples as the working electrode and a platinum foil as the counter electrode. The distance between the

two electrodes was 20 mm. Anodization was carried out at a constant time of 1 h for different voltage. Besides, anodization was carried out at a constant DC voltage of 30 V for 1 h, 2 h, 4 h, respectively. After anodization, all samples were immediately cleaned by distilled water and dried in cool air to remove the electrolyte and the undesired disordered layers on their surfaces. To get anatase crystalline phase nanotube layers, the nanotubes were then sintered at 450 for 3 h.

13684

Ceramics International 43 (2017) 13683–13688

J. Li et al.

Fig. 2. Characterization of TiO2-Mg NTs for different time at 30.0 V. (a) the surface SEM images of the anodized samples; (b) histogram of the nanotube diameter; (c) the cross-sectional images of the anodized samples.

properties of the coatings. Potentiodynamic polarization and EIS were performed in SBF at room temperature. The reagent grade chemicals of SBF were used as follows: 8.035 g/L NaCl, 0.355 g/L NaHCO3, 0.225 g/L KCl, 0.231 g/L K2HPO4·3H2O, 0.311 g/L MgCl2·6H2O, 1.0 M HCl (39 ml), 0.292 g/L CaCl2, and 0.072 g/L Na2SO4 in deionized water and the pH was adjusted to 7.4 using 6.118 g/L trishydroxymethyl aminomethane (Tris) and 1.0 M HCl at 37 °C. Electrochemical measurements were measured by conventional three-electrode system, where a saturated calomel electrode (SCE) as the reference electrode, a platinum foil as the counter electrode and the samples as the working electrode. The OCP was measured at the scan rate of 1 mV/s to obtain a steady-state OCP value. The EIS was carried out at recorded at OCP by applying a 10 mV sinusoidal potential through a frequency domain from 100 kHz down to 10 mHz with an excitation amplitude of 10 mV. The potentiodynamic polarization curves were scanned from −0.5 to 1.5 V at a scanning rate of 0.5 mV/s. Three samples of every surface were conducted for the corrosion test. The data is fitted by ZSimpWin software.

Fig. 3. The OCP values of AZ91D, Ti-Mg and TiO2-Mg in SBF for 30 min.

2.3. Sample characterization

3. Result and discussion

The surface and cross-sectional morphologies of the samples were measured by using a field-emission scanning electron microscope (FESEM, JSM-7001F, JEOL).

Anodization voltage plays a vital role in controlling the surface morphology of TiO2-Mg NTs. As shown in Fig. 1(a), the surface morphology of the sample anodized at 20 V gives rise to a small amount of floccule. 30 V and 40 V is similar with aligned and uniform structure. Increasing the voltage from 20 to 40 V results in an increase in the NT diameter from 40 to 50 nm at 20 V to 60–70 nm at 40 V. The change can be attributed to the increasing the transmission rate of ions over the oxide layer accelerating the oxidation rate of the metal [19]. With regard to 50 V, anodization produces an irregular nanoporous structure with a thickness of 70–110 nm. Compared to TiO2 NTs, the diameter of TiO2-Mg

2.4. Electrochemical measurements To avoid undesirable ohmic effects, all specimens were carefully glued to the copper wire with conductive paper. And then these specimens mounted with cold epoxy resin. The PMC-2000 electrochemical workshop was used to evaluate the corrosion resistance

13685

Ceramics International 43 (2017) 13683–13688

J. Li et al.

Fig. 4. The experimental and simulated Nyquist (a) and bode plots (b, c) of AZ91D, Ti-Mg and TiO2-Mg at respective OCP in SBF.

Fig. 5. Equivalent circuit models for the surfaces of different samples: (a) AZ91D; (b) TiMg; (c) TiO2-Mg.

NTs decreases slightly at same voltage. The current density-time curves in Fig. 1(e) also exhibit two distinct tendencies at 30 V. The current density drops rapidly initially and then decreases gradually until reaching a relatively stable value. The current density values of TiO2 NTs is bigger than that of TiO2-Mg NTs. The bigger current density can promote the ion mobility thus accelerating F¯ etching of the oxide layer, further forming the bigger diameter. The other important factor that influences the anodic behavior of Ti-Mg is the applied time. Fig. 2(a) shows the SEM images of the top surface morphologies of the prepared TiO2-Mg NTs obtained with various anodization times. It can be seen that the ordered and homogeneous nanotubular structure on Ti-Mg films has been formed, and the diameter of NTs obviously increases with anodization time increased, changing from 50 to 60 nm at 1 h to 60–80 nm at 4 h (Fig. 2(b)). The cross-sectional SEM exhibits that the length of nanotubes also increases with time (Fig. 2(c)). It can be inferred that the growth rate of NTs is gradually reduced. With the increase of the NT film's thickness, the diffusion of F- became slow resulting in the decrease of etching rate.

The OCP can monitor the chemical stability and estimate the tendency of corrosion behavior. Fig. 3 shows the OCP values for AZ91D, Ti-Mg and TiO2-Mg in SBF. The OCP of AZ91D rose slowly but then gradually slowed reaching a relatively stable value. After the AZ91D electrode was immersed into the electrolyte, an Mg oxide film began to grow on the sample surface. The relatively stable value is attributed to the slow and regular reactions in the electrolyte. However, the OCP values of Ti-Mg and TiO2-Mg were decreased slightly and then reached respectively a stable value around −1.39 V, −1.40 V, which show the larger noble potential shifts. For Ti-Mg, this may result from the non-uniform composition on the surface and the formation of oxide layer on the surface because of the dissolved oxygen. And the thickness of surface Ti oxides film increase with the increase of immersion time, the potential stable can reach the stable state when a stable thickness of oxide film. For TiO2-Mg, the slight decrease in the values may be on account that the electrolyte passes through the pores in the outer layer and enters into the interface between substrate and nanoporous layer [20,21]. The ceramic TiO2 can efficiently inhibit electron transfer so the corrosion resistance increase. Amongst the tested samples, the AZ91D has the lowest voltage value and the longest time to reach stable value, indicating that the great corrosion tendency. The representative Nyquist (a) and Bode plots (b, c) at respective OCP obtained from AZ91D, Ti-Mg and TiO2-Mg in SBF are shown in Fig. 4. For AZ91D, the Niquist graph was composed of a big capacitive loop at high frequency, a small capacitive loop at medium frequency and an inductive loop at low frequencies. The higher frequency capacitive behavior results from charge transfer indicating that the electrolyte penetration can generate corresponding resistance in the faradic process parallel to the double layer capacitance. The middle frequency capacitive is associated with mass transport stemming from

13686

Ceramics International 43 (2017) 13683–13688

100.7 ± 3.12 – – 395.8 ± 8.42 – – 160.7 ± 2.23 1.8 × 103 ± 86.9 – 0.87 ± 0.02 0.89 ± 0.01 – 1.07 × 103 ± 1.27 1.26 × 10−1 ± 0.002 – 147.5 ± 1.3 121.2 ± 0.9 134.2 ± 1.7 AZ91D Ti-Mg TiO2-Mg

15.2 ± 1.3 2.57 × 10−5 ± 2.3 × 10−7 5.36 × 10−5 ± 1.3 × 10−8

0.88 ± 0.01 0.92 ± 0.02 0.93 ± 0.00

163.7 ± 6.29 2.6 × 109 ± 10629 2.74 × 109 ± 13021

L (H) R2 (Ω cm2) n2 n1 CPE1 (μF / cm2) RS (Ω cm2) Samples

Table 1 Electrochemical parameters obtained from EIS spectra using the equivalent circuits in Fig. 5.

R1 (Ω cm2)

CPE2 (μF / cm2)

R3 (Ω cm2)

J. Li et al.

Fig. 6. Potentiodynamic polarization behavior of AZ91D, Ti-Mg and TiO2-Mg in SBF solutions at a scan rate of 1 mV/s.

the metal ion diffusion. And the low inductive loop can be related to the adsorption indicating nucleation of the corrosion pits [6,22]. The shape of Nyquist plots for Ti-Mg show two capacitive loops over the entire frequency range, whereas TiO2-Mg shows just only one capacitive loop. In addition, the radii of Ti-Mg and TiO2-Mg are bigger than that of AZ91D suggesting the improved corrosion resistance. However, from the beginning of media frequency, the radius of Ti-Mg suddenly decreases indicating the low stability of Ti film. It can be inferred that the corrosion resistance of TiO2-Mg is better than that of Ti-Mg in the long term. From Bode impedance and Phase angle plots in Fig. 4b and c, Ti-Mg and TiO2-Mg show larger impedance magnitude (|Z|) and phase angle over entire frequency range compared to the untreated substrate. In the high and media frequency region, for Ti-Mg and TiO2-Mg, the Bode impedance show a linear relationship with a slope of −0.83 and phase angle approach to −90°, suggesting obvious capacitive behavior of the electrode/electrolyte interface [23]. The higher phase angle shows the more capacitive behavior showing the formation of oxide layer on the surface of Ti-Mg. For TiO2-Mg, This phenomenon suggests that only a small amount of corrosion solution can enter the nanotube through the oxidized/product layer, further indicating that the oxidized/product layer is compact and uniform. In addition, the maximum phase angle can imply low corrosion tendency and rate [24]. In the low frequency region, Ti-Mg and TiO2-Mg have significantly improved corrosion protection compared to the AZ91D evidenced by the |Z| of higher than corresponding value of AZ91D by 8 orders of magnitude. Therefore, the significant corrosion improvement of Ti-Mg and TiO2Mg in SBF can be attributed to the formation of passive film, but the film of Ti-Mg is unstable resulting in the poor long-term corrosion. Compared to TiO2-Mg, Ti film have not been completely oxidized at 30 V for 1 h, so TiO2 NTs can tightly adhere to the Ti film. Thus, it can be inferred that the corrosion solution firstly enter into nanotubes and crevice and then cause two different effects, one destroys the construction of TiO2 NTs to form the corrosion layer further preventing the penetration of solution. And one combines with unoxidized Ti film to form titanium oxide protective film. These two protective film can effectively enhance the corrosion resistance of substrate. According to surface morphology and above analysis, the EIS spectra are fitted using an electrical equivalent circuit model, Rs(CPE1(R1(CPE2R2)(L1R3))), Rs(CPE1(R1(CPE2R2))), Rs(CPE1R1) as shown in Fig. 5, respectively. In the equivalent circuit, Rs is the solution resistance between the samples and reference electrodes; R1 is the resistance of corrosion productions for AZ91D and Ti-Mg, namely, Mg(OH)2 porous layer and TiO2 layer. For TiO2-Mg, R1 is the resistance of TiO2 NTs. CPE1 is the corresponding capacitance. R2 is the polarization resistance at electrode/electrolyte interface paralleled with CPE2, R3 13687

Ceramics International 43 (2017) 13683–13688

J. Li et al.

References

Table 2 Polarization data of Mg, Ti-Mg and TiO2-Mg in SBF solutions. Samples

Ecorr (V)

Icorr (A/cm2)

AZ91D Ti-Mg TiO2-Mg

−1.66 ± 5.6 × 10−3 −1.34 ± 2.1 × 10−2 −1.29 ± 1.2 × 10−2

5.89 × 10−5 ± 3.2 × 10−6 1.52 × 10−4 ± 8.7 × 10−7 4.82 × 10−5 ± 4.6 × 10−7

Data are shown as mean ± standard deviation (n = 3).

is the resistance of local corrosion. L1 is inductor. The circuit elements calculated from the fitting results are summarized in Table 1. The electrolyte resistance (denote as Rs) of all specimens is about 130 Ω cm2 because of the fixed positions of the working and reference electrode. Compared to the AZ91D, the values of R1 for Ti-Mg and TiO2-Mg increase more than orders of magnitude due to the compact and uniform corrosion layer on the surface, which can protect effectively the substrate against the corrosion solution. Besides, much lower values of CPE1 for Ti-Mg and TiO2-Mg show higher stability property of oxide film. And their n1 range from 0.88 to 0.93 indicating they approach to the extent of pure capacitances. The largest value of R1 for TiO2-Mg shows the formation of stable and uniform protective film. Potentiodynamic polarization curves for AZ91D, Ti-Mg and TiO2Mg are shown in Fig. 6. The curves obviously show an enhanced corrosion potential and a reduced corrosion current density for Ti-Mg and TiO2-Mg. Table 2 shows the corrosion potential (Ecorr) and corrosion current density (Icorr) calculated using Tafel extrapolation. The increasing of Ecorr indicates that Ti-Mg and TiO2-Mg are more cathodic compared to AZ91D, suggesting Ti films can effectively retard electrons transfer thus leading to better corrosion resistant. The corrosion resistance of TiO2-Mg is slightly better than that of Ti-Mg, this can attribute to the inherent and thicker TiO2 layer passive barrier on TiO2-Mg surface [17,25]. In addition, AZ91D and Ti-Mg exist obvious pitting platform suggesting the inhomogeneity of Ti film. The polarization results are consistent with those obtained by Nyquist and Bode impedance studies.

4. Conclusion TiO2-Mg NTs with different size has been successfully fabricated on Ti-Mg realizing by anodizing magnetron-sputtered Ti coatings on AZ91D through varying anodization voltages and time. The Ti-Mg with the length of 60–70 nm anodized at 30 V and 1 h has distinct and uniform nanotubular structure. The corrosion of TiO2-Mg NTs has effectively improved compared to AZ91D. Overall, anodization on deposition of Ti by magnetron-sputtering is a promising method of decreasing the fast degradation rate of Mg alloys.

Acknowledgment This work was supported by the National Natural Science Foundation of China (51671140 and 31400815), the Natural Science Foundation of Shanxi Province (2015021063), the Qualified Personnel Foundation of Taiyuan University of Technology (QPFT) (tyutrc2011257a) and the Research Project Supported by Shanxi Scholarship Council of China (2015-034).

[1] M. Razavi, M. Fathi, O. Savabi, S. Mohammad Razavi, B. Hashemi Beni, D. Vashaee, L. Tayebi, Surface modification of magnesium alloy implants by nanostructured bredigite coating, Mater. Lett. 113 (2013) 174–178. [2] J.W. Guo, S.Y. Sun, Y.M. Wang, Y. Zhou, D.Q. Wei, D.C. Jia, Hydrothermal biomimetic modification of microarc oxidized magnesium alloy for enhanced corrosion resistance and deposition behaviors in SBF, Surf. Coat. Technol. 269 (2015) 183–190. [3] L.G. Bland, K. Gusieva, J.R. Scully, Effect of crystallographic orientation on the corrosion of magnesium: comparison of film forming and bare crystal facets using electrochemical impedance and Raman spectroscopy, Electrochim. Acta 227 (2017) 136–151. [4] M.I. Jamesh, G. Wu, Y. Zhao, D.R. McKenzie, M.M.M. Bilek, P.K. Chu, Effects of zirconium and oxygen plasma ion implantation on the corrosion behavior of ZK60 Mg alloy in simulated body fluids, Corros. Sci. 82 (2014) 7–26. [5] S. Agarwal, J. Curtin, B. Duffy, S. Jaiswal, Biodegradable magnesium alloys for orthopaedic applications: a review on corrosion, biocompatibility and surface modifications, Mater. Sci. Eng. C 68 (2016) 948–963. [6] Y. Xin, C. Liu, W. Zhang, K. Huo, G. Tang, X. Tian, P.K. Chu, Corrosion resistance of ZrO2-Zr-coated biodegradable surgical magnesium alloy, J. Mater. Res. 23 (2011) 312–319. [7] Y. Li, L. Wang, D. Zhang, L. Shen, Influence of bias voltage on the formation and properties of iron-based nitrides produced by plasma nitriding, J. Alloy. Compd. 497 (2010) 285–289. [8] G. Wu, J.M. Ibrahim, P.K. Chu, Surface design of biodegradable magnesium alloys — a review, Surf. Coat. Technol. 233 (2013) 2–12. [9] M. Ratova, P.J. Kelly, G.T. West, L. Tosheva, M. Edge, Reactive magnetron sputtering deposition of bismuth tungstate onto titania nanoparticles for enhancing visible light photocatalytic activity, Appl. Surf. Sci. 392 (2017) 590–597. [10] M.R. Souza, K.M. Reyes, N.T.C. Oliveira, N.K. Kuromoto, C.E.B. Marino, Bioactivity of self-organized TiO2 nanotubes used as surface treatment on Ti biomaterials, Mater. Res. Express 3 (2016) 035401. [11] N. Wang, H. Li, W. Lu, J. Li, J. Wang, Z. Zhang, Y. Liu, Effects of TiO2 nanotubes with different diameters on gene expression and osseointegration of implants in minipigs, Biomaterials 32 (2011) 6900–6911. [12] D. Khudhair, A. Bhatti, Y. Li, H.A. Hamedani, H. Garmestani, P. Hodgson, S. Nahavandi, Anodization parameters influencing the morphology and electrical properties of TiO2 nanotubes for living cell interfacing and investigations, Mater. Sci. Eng. C 59 (2016) 1125–1142. [13] M. Zong, L. Bai, Y. Liu, X. Wang, X. Zhang, X. Huang, R. Hang, B. Tang, Antibacterial ability and angiogenic activity of Cu-Ti-O nanotube arrays, Mater. Sci. Eng. C 71 (2017) 93–99. [14] R. Hang, A. Gao, X. Huang, X. Wang, X. Zhang, L. Qin, B. Tang, Antibacterial activity and cytocompatibility of Cu-Ti-O nanotubes, J. Biomed. Mater. Res. A 102 (2014) 1850–1858. [15] D. Wang, Y. Liu, B. Yu, F. Zhou, W. Liu, TiO2 nanotubes with tunable morphology, diameter, and length: synthesis and photo-electrical/catalytic performance, Chem. Mater. 21 (2009) 1198–1206. [16] J. Xiong, X. Wang, Y. Li, P.D. Hodgson, Interfacial chemistry and adhesion between titanium dioxide nanotube layers and titanium substrates, J. Phys. Chem. C 115 (2011) 4768–4772. [17] W.Q. Yu, J. Qiu, L. Xu, F.Q. Zhang, Corrosion behaviors of TiO2 nanotube layers on titanium in Hank's solution, Biomed. Mater. 4 (2009) 065012. [18] K. Lee, A. Mazare, P. Schmuki, One-dimensional titanium dioxide nanomaterials: nanotubes, Chem. Rev. 114 (2014) 9385–9454. [19] R. Zhang, H. Wu, J. Ni, C. Zhao, Y. Chen, C. Zheng, X. Zhang, Guided proliferation and bone-forming functionality on highly ordered large diameter TiO2 nanotube arrays, Mater. Sci. Eng. C 53 (2015) 272–279. [20] H. Duan, K. Du, C. Yan, F. Wang, Electrochemical corrosion behavior of composite coatings of sealed MAO film on magnesium alloy AZ91D, Electrochim. Acta 51 (2006) 2898–2908. [21] W. Jin, G. Wu, H. Feng, W. Wang, X. Zhang, P.K. Chu, Improvement of corrosion resistance and biocompatibility of rare-earth WE43 magnesium alloy by neodymium self-ion implantation, Corros. Sci. 94 (2015) 142–155. [22] Q. Liu, D. Chen, Z. Kang, One-step electrodeposition process to fabricate corrosionresistant superhydrophobic surface on magnesium alloy, ACS Appl. Mater. Interfaces 7 (2015) 1859–1867. [23] X. Zhang, H. Wang, J. Li, X. He, R. Hang, X. Huang, L. Tian, B. Tang, Corrosion behavior of Zn-incorporated antibacterial TiO2 porous coating on titanium, Ceram. Int. 42 (2016) 17095–17100. [24] W. Jin, G. Wang, Z. Lin, H. Feng, W. Li, X. Peng, A.M. Qasim, P.K. Chu, Corrosion resistance and cytocompatibility of tantalum-surface-functionalized biomedical ZK60 Mg alloy, Corros. Sci. 114 (2017) 45–56. [25] N. Figueira, T.M. Silva, M.J. Carmezim, J.C.S. Fernandes, Corrosion behaviour of NiTi alloy, Electrochim. Acta 54 (2009) 921–926.

13688